1. Field of the Invention
The present invention relates to methods and compositions for the priming of a specific cytotoxic T-lymphocyte (CTL) response, and to the use of these methods in the identification of candidate substances, typically polypeptides, for use in the preparation of CTL vaccines having the ability to prime an in vivo CTL response. This technology will have particular application in the prevention and treatment of viral diseases, such as AIDS, herpes, influenza, feline leukemia, and the like.
2. Description of the Related Art
AIDS was first recognized in the United States in 1981; the number of cases has been increasing at a dramatic pace since then. Since 1978 more than 2.4 million AIDS infections have been reported in the United States, alone (Rees, 1987). Once significant immunosuppressive symptoms appear in an infected individual, the expected outcome of the infection is death. There is currently no known treatment that can indefinitely delay or prevent the fatal consequences of the disease. Although the disease first manifested itself in homosexual or bisexual males and intravenous drug abusers, it has now spread to others by means such as intimate sexual contact with or receipt of blood products from a carrier of the virus.
The causative agent, associated with AIDS has been identified as a group of closely related retroviruses commonly known as Human T Cell Lymphotrophic Virus-type III (HTLV-III), Lymphadenopathy Viruses (LAV), AIDS-Related Viruses (ARV), or more recently named Human Immunodeficiency Virus (HIV). These viruses will be collectively referred to herein for convenience as HIV.
Like other retroviruses, HIV has RNA as its genetic material. When the virus enters the host cell, a viral enzyme known as reverse transcriptase copies the viral RNA into a double stranded DNA. The viral DNA migrates to the nucleus of the cell where it serves as a template for additional copies of viral RNA which can then be assembled into new viral particles. The viral RNA can also serve as messenger RNA for certain viral proteins [either the viral core proteins (known as p18, p24 and p13)] or the reverse transcriptase, or be "spliced" into specific viral messenger RNAs necessary to produce several other viral proteins including two glycosylated structural proteins known as gp41 and gp120 which are inserted in the outer membrane of the virus (Wain-Hobson et al., 1985). A recent study has shown that purified gp120 induces antibody in the goat, horse and rhesus monkey that neutralizes HIV in lab tests (Robey et al., 1986).
Vaccines have been used for many years to prevent infections caused by agents such as viruses. The general approach has been to inject healthy individuals with, for example, a killed or modified virus preparation in order to prime the individual's immune systems to mount an assault on the infecting virus. Recent advances in recombinant DNA technology have allowed safer methods of vaccination that involve use of exposed viral components produced by microbial systems. After sufficient purification, the viral component, for example a protein subunit, is administered as a vaccine in a suitable vehicle and/or an adjuvant. The latter stimulates the host's system in a way that improves the immune response to the viral subunit.
Another potential method of making a vaccine is by using chemically synthesized peptide fragments of a viral protein subunit. This method has several advantages over the other methods of producing vaccines, including purity of the product, reproducibility and specificity of the immune response.
Surface antigens of an infecting virus can elicit T cell and B cell responses. From the work of Milich and coworkers (Milich et al., 1986; Milich & McLachlan, 1986) it is clear that some regions of a protein's peptide chain can possess either T cell or B cell epitopes. These epitopes are frequently distinct from each other and can comprise different peptide sequences. Other examples include the work of Maizel et al., (1980) for hen egg-white lysozyme, and Senyk et al., (1971) for glucagon. Thus, short stretches of a protein sequence can elicit a T cell response but not a B cell response. A more complete review of these and other observations pertinent to this point is included in the work of Livingstone & Fathman (1987).
A short peptide region within the surface protein of infectious Hepatitis B virus has been shown to elicit only a T cell response in mice (Milich et al., 1986). Specifically, a synthetic peptide, whose sequence is derived from amino acids numbered 120-132 located within the pre-S(2) domain of the Hepatitis B surface antigen gene, elicited a very strong T cell priming response to the peptide but stimulated only a very weak antibody response. In other words, mice mounted a poor antibody response to that peptide, but the T cells of immunized mice were efficiently primed (i.e. activated) to recognize that peptide as measured in T cell proliferation assays (Milich et al., 1986). The low level of the antibody produced by mice immunized with this peptide did not bind to the native viral surface antigen.
In contrast to the above-described results, a second peptide sequence (amino acids 132-145) elicited a very weak T-cell response in mice (Milich et al., 1986). This second peptide did, however, efficiently bind antibody raised against it under conditions where a T cell epitope is provided.
Mice were also immunized with a longer peptide made up of both of the above-mentioned T- and B-active peptide sequences. In this case, high titers of antibody were produced against the B site peptide but not the T site peptide. The combination of both T- and B-sites within one peptide should stimulate both T and B cell responses, as measured by producing a specific antibody to the B cell epitope of the peptide chain. Synthetic peptide antigens may be constructed to produce two types of immune responses: T-cell only and T cell combined with a B cell response.
Cellular immune responses provide a major mechanism for reducing the growth of virus-infected cells (Doherty et al., 1985). A report by Earl et al., (1986) demonstrated T-lymphocyte priming and protection against the Friend virus (a retrovirus)-induced mouse leukemia by a viral surface protein vaccine. Direct evidence for the role of a subset of T-lymphocytes (OKT8/LEU2 positive) in suppressing HIV growth in vitro was recently obtained by Walker et al. (1986). This study further demonstrated that, after depletion of CD8.sup.+ T-lymphocytes from the blood of HIV-infected individuals, large quantities of HIV were isolated from peripheral blood mononuclear cells of four of seven asymptomatic, seropositive homosexual men who were initially virus-negative or had very low levels of virus. Thus, the CD8.sup.+ cytotoxic T-lymphocytes (CTLs) may play a role in virus infected individuals to prevent HIV replication and disease progression.
The concept of identifying T-cell epitopes in proteins for inclusion in potential vaccine candidates has gained importance as a result of the demonstration by Townsend et al. (1986) that CTL epitopes of influenza nucleoprotein can be defined by short synthetic peptides. However, to date there are only three documented cases (Deres et al., 1989; Aichele et al., 1990; Kast et al., 1991) that describe the use of synthetic peptides in the in vivo priming of CTLs, these relate to influenza, Sendai and lymphocyte choriomeningitis viruses. In each of the above cases, the immunization protocols are cumbersome, require either modifications of peptides or many immunizations to be carried out to demonstrate CTLS, and do not lend themselves to the rapid screening of a large number of candidate substances. For example, the method of Aichele and colleagues (1990) involves three immunizations at one week intervals by the subcutaneous route, and takes four weeks before potential CTLs are obtained for assaying.
Candidate CTL epitopes in both structural and regulatory HIV proteins have been proposed (Takahashi et al., 1988; Nixon et al., 1988) but none of these have been shown to be capable of inducing virus-specific CTLs in vivo (Berzofsky, 1991). For example, although the peptide RIQRGPGRAFVTIGK has been identified as a CTL epitope (Takahashi et al., 1988), in these studies the in vivo induction of RIQRGPGRAFVTIGK-specific CTLs was accomplished by infecting Balb/c mice with recombinant vaccinia virus expressing HIV env proteins (Takahashi et al., 1988) and attempts at immunization with free peptide have been unsuccessful (Berzofsky, J. A., 1991).
There is thus clearly a need for techniques for the rapid identification of CTL-reactive epitopes that have the ability to prime a specific T-cell response. Previous methods have suffered a number of drawbacks, most notably, their requirement for multiple injections of material to be tested, a wait of up to 3 weeks or longer to determine whether the substance had a positive effect on CTL response, and the general need to include a modifier with the substance being tested in order to elicit a response. Clearly, a rapid method for the delineation of peptides with an in vivo CTL inducing capacity is of vital importance in the design of preventative and therapeutic strategies in relation to disease such as AIDS.